The identification of lower-penetrance susceptibility alleles for common cancers in case-control association studies has been disappointing thus far. This new initiative aims to concentrate our efforts to identify such susceptibility genes in populations and subjects likely to be enriched for inherited factors. This is in contrast to population-based epidemiologic studies in which cases are not specifically selected to be enriched for genetic susceptibility factors. The emphasis in both potential projects are non-BRCA1/2 related genes predisposing to breast or ovarian cancer. Ovarian cancer susceptibility alleles will be sought by comparing allele frequencies, both of candidate genes and eventually whole-genome scans, in ovarian cancer cases diagnosed before age 50 or those with a family history of ovarian cancer with unaffected control individuals. The second initiative we hope to begin is to study sister pairs affected with breast cancer among inbred populations, specifically among Bedouin women from Jordan. Because there is evidence that recessive susceptibility alleles contribute to a significant portion on non-BRCA1/2-related breast cancer, this may be an ideal population in which to map these genes.